Many elements have chemically identical isotopes, which vary only in the number of their neutrons. These isotopes of a single element are often co-produced in a single reaction. The isotopes of a single element do vary, particularly in their radioactivity, and therefore purification or separation of the isotopes leads to an improved product. However, these isotopes, being chemically identical, are difficult to separate.
By way of example, iodine-123 is a close-to-ideal radioactive isotope used in nuclear medicine. It permits diagnostic tomography of a patient's brain to allow a physician to distinguish between multiple infarct dementia, associated with a series of strokes, and Alzheimer's Disease. These two diseases, requiring quite different treatment, have, prior to diagnostic tomography, given only obscure and conflicting symptoms.
Pure iodine-123 is close-to-ideal because it has a 13 hour half life allowing shipment nationwide and yet decaying sufficiently rapidly so that the patient does not receive an excessive dosage of radiation. Pure iodine-123 emits a single characteristic energy, 159 keV, and practically no other radiation. Pure iodine-123 is well tolerated by the human organism and is readily compounded as a label into many biochemical agents without disrupting their action.
The principal drawback to the use of pure iodine-123 in nuclear medicine is that pure iodine-123 is expensive to produce. Most of the iodine-123 which is available is not pure. It is produced by a 20 to 40 MeV cyclotron or linear accelerator. The high energy protons from the cyclotron bombard a target enriched in tellurium-124, whose atomic number is 52 compared with iodine whose atomic number is 53. The high energy proton principally dislodges two neutrons to produce iodine-123 in the reaction: EQU Te-124 (p,2n).fwdarw.I-123
Some of the bombarding protons dislodge only one neutron. In addition, the target contains residual amounts of tellurium-125 and tellurium-126. The result is that about 3% of iodine-124 and iodine-125 is co-produced in the side reactions: EQU Te-124 (p,n).fwdarw.I-124 EQU Te-125 (p,n).fwdarw.I-125 EQU Te-126 (p,2n).fwdarw.I-125
These two isotopes of iodine, even a few percent, are each unwanted in nuclear medicine for different reasons. Iodine-124 emits other radiation which degrades the output of diagnostic tomography, making it a less precise test. The radiation from iodine-125 is soft, i.e., X-rays which are absorbed within the patient, and therefore does not degrade the output of diagnostic tomography. However, iodine-125 has a half life of 60 days, thus substantially increasing the internal radiation dosage to the patient.
Two alternative processes exist to produce pure iodine-123. The first one uses a 20-40 MeV cyclotron in which the high energy proton bombards a target of xenon-124 and dislodges two neutrons to produce cesium-123. This then decays in 8 minutes to xenon-123, which decays in 2 hours to iodine-123, according to the reaction: EQU Xe124 (p,2n).fwdarw.Cs-123 (8 min).fwdarw.Xe 123 (2 hrs).fwdarw.I-123
This reaction produces high purity iodine-123, but does so very expensively because xenon-124 is a very rare isotope. The process was, until recently, therefore limited to research applications.
The second process uses a 70 MeV cyclotron with an iodine 127 target. The high energy proton can dislodge 5 neutrons to produce xenon-123, which decays in 2 hours to a solid iodine-123 according to the reaction: EQU I-127 (p,5n).fwdarw.Xe-123 (2 hr).fwdarw.I-123
In a first side reaction the proton dislodges 4 neutrons to produce a gaseous xenon-124, according to the reaction: EQU I-127 (p,4n).fwdarw.Xe-124 (stable)
which is a stable gas, and further separates from the solid iodine-123.
Another side reaction occurs. In this, the proton dislodges three neutrons to produce xenon-125, which decays to iodine-125, according to the reaction: EQU I-127 (p,3n).fwdarw.Xe-125.fwdarw.I-125
About 0.2% of iodine-125 is co-produced and is a contaminant.
To date, all 70 MeV cyclotrons have been built for research, not commercial applications, due to the expense of building and operating them. One 70 MeV cyclotron should become operational in the United States in 1986, which will be the first one dedicated to the commercial production of radioactive isotopes for nuclear medicine.
This low recovery rate is acceptable in an analytical or research magnetic mass spectrometer. This technology becomes uneconomic, however, for commercial applications, including separating radioactive isotopes for nuclear medicine.
A technology to purify or separate a polyisotopic mixture has existed since the 1920's. This is the magnetic mass spectrometer or mass analyzer, which achieves a physical separation according to mass by differentially deflecting ionized isotopes of different atomic weight, the lighter isotopes being deflected more than the heavier isotopes.
The first step in a magnetic mass spectrometer is to ionize the isotopes in an ion source. At most only a small portion of the isotopes can be ionized and the remainder of the isotopes escape into the vacuum chamber with no charge and are therefore neither accelerated by the electrode nor deflected by the magnet. Recoveries of 1% to 15% of the desired isotope are typical.
These elements of the present invention appear in innumerable prior art publications, and applicant will not attempt to separately identify what may be the closest prior art.
The applicant has found only two items in the prior art pertinent to the present invention:
Simmons, U.S. Pat. No. 2,533,966, issued Dec. 12, 1950, discloses a novel method to accelerate ions, a moving magnetic field. Simmons also discloses an ion source 17, having two feed pipes for ions, 13 and 14, and two exit pipes 19 and 22. Pipe or casing 22 contains a focusing field 45 for the ionized isotopes to exit into the magnetic field of force 48 of a mass spectrometer. At the other end of the ion source is a grid 20 which repels the ionized isotopes but allows the un-ionized isotopes to enter pipe or conduit 19. These isotopes, supplemented by the feed stock, are recirculated to the ion source through pipes 13 and 14. There is no means to prevent un-ionized isotopes from escaping at the larger opening 21 nor any recirculation of any un-ionized isotopes which escape the ion source 17 at opening 21.
There are crucial differences between the Simmons disclosure and the present invention. First, the present invention recirculates the uncharged isotopes which have left the ionizer chamber with the charged ions, while Simmons withdraws uncharged isotopes from the ionizer and merely recirculates them to the ionizer with no apparent means for increasing the recovery rate thereby. Secondly, the present invention recirculates by first adsorbing and then desorbing the iodine isotope on a glass surface. Simmons recirculates the isotope in the same gaseous state in which it left the ionizer.
Meunier, et al., Nuclear Instruments and Methods 139 (1976) 101-104 discloses a "closed loop circuit" used on two research machines for isotope separation near Paris, France. The machine has a conventional ion source, aperture, vacuum chamber and magnetic mass spectrometer.
Spaced along the vacuum chamber are a series of side chambers, each containing a vacuum pump for the un-ionized isotopes. There is a cascade series of pumps, including oil diffusion pumps, which raise the pressure from 10.sup.-5 Torr to atmospheric pressure. The recovered isotopes are mixed with feed stock and reintroduced to the ion source. The overall recovery rate ranges from 11 to 32%, perhaps a four-fold improvement over conventional, non-recovery isotope separators. This system is limited to inert gases which will not bond to the metal pump parts.
The present invention is an internal recycle under high vacuum, while Meunier is an external recycle utilizing several cascaded pumps to raise the isotope to atmospheric pressure, then reintroduce it to the ionizer under vacuum. The present invention has a recycle loop which is clean, non-reactive, and therefore suitable for iodine. It is non-contaminating, simple, compact and relatively cheap with a recovery rate which may approach 100%. Meunier's recycle loop introduces impurities, is suitable only for inert gases, is complicated, bulky, and expensive. The best utilization reported is 32%.